Visual and laser data integrated road detection method

Abstract

The invention discloses a visual and laser data integrated road detection method, comprising the following steps: S1) acquiring image data through a camera; using super-pixel segmentation to segment the image; S2) using the Adaboost principle to extract characteristic vectors from the image data acquired by the camera; S3) using an online learning based method to detect a road area; S4) using a morphological method to optimize the detection result; S5) using the integration of laser and visual data to map the visual detection result with the laser data; correcting the detection result; and S6) constructing a three-dimensional road area in the laser data; and fitting a three-dimensional road boundary. The classifier of the invention can update along with the change of the environment so that it can adapt to the changes in weather, seasons and light rays well and that the robustness of the algorithm can be ensured. The integration technology of laser and visual data realizes the seamless integration of a road detection result with the navigator of a driveless car.

Description

technical field [0001] The invention specifically relates to a road detection method of visual laser data fusion. Background technique [0002] As a hot field of scientific and technological development today, intelligent technology has benefited all aspects of human beings. As an outstanding representative of intelligent technology, robotics technology has gradually matured and moved toward commercialization after more than 50 years of development. [0003] Unmanned vehicles are a typical mobile robot, and road detection, as a research hotspot in the field of unmanned vehicles, has produced many related results in the past decade. Although there are many proposals, model-based and feature-based methods, supervised and unsupervised methods, there are two key issues that are quite challenging to compare the core. [0004] (1) There are many types of roads in the unstructured environment, the characteristics are unstable, and it is easy to change greatly due to factors such ...

AI Technical Summary

Problems solved by technology

Although there are many proposals, model-based and feature-based methods, supervised and unsupervised methods, there are two key issues that are quite challenging.

[0004] (1) There are many types of roads in the unstructured environment, the characteristics are unstable, and it is easy to change greatly due to factors such as light, weather, and seasons. Therefore, driverless vehicles for unstructured environments must have the ability to adapt to environmental changes. ability

[0005] (2) The most commonly used sensor for road detection is a camera, so the detection result is usually the road area in the image, but the image itself does not have three-dimensional features, so there are only areas in the image, and the result is difficult to apply to unmanned vehicles. In actual navigation needs, there are also algorithms that use laser sensors as detection methods. The detection results can be directly used for navigation and path planning. However, the characteristics of laser data are relatively few, so it is difficult to improve detection accuracy and adaptability.

Images